Limitations of first-phase insulin response to evaluate insulin secretion in children

The first-phase insulin response (FPIR) is a widely used method to evaluate beta-cell function during the prediabetic phase of evolving type 1 diabetes mellitus (DM). The aim of the present study was to evaluate the influence of clinical and methodological variables on FPIR in normal children and adolescents. Children and adolescents who were first-degree relatives of those with type 1 DM and healthy young adults were studied. All subjects were islet cell antibody-negative. A FPIR test was performed on all subjects while fasting. Insulin samples were drawn at 0, 1, 2, 3, 4, 5, 6, 8, and 10 min after 0.3 g/kg of dextrose. FPIR(1-10) was calculated as the area under the FPIR curve corrected for baseline. Eighty-five subjects aged 4-22 yr were studied, 43 of whom were pre-pubertal, 24 pubertal, and 18 post-pubertal. FPIR(1-10) values were lower in the pre-pubertal group when compared to either the pubertal and post-pubertal groups (415 [179-965, 2SD], 756 [256-2 223] and 684 [235-1 180] mU/L, respectively; p<0.05). Obese subjects had a higher FPIR than non-obese subjects (856 vs. 520 mU/L; p<0.005). Despite correcting for the influence of puberty and obesity, there remained considerable unaccounted variability in FPIR(1-10) (R=0.46). Further variables found to influence FPIR(1-10) were: fasting insulin level (r(2)=0.49); weight for length index (r(2)=0.38); peak blood glucose level (r(2)=0.38, all p<0.001); and pre-pubertal age (r(2)=0.20, p<0.05).

FPIR exhibited wide inter-subject variability and was influenced by a number of clinical and methodological factors that make interpretation more difficult without more specifically defined standards.